chapter 2 literature survey -...
TRANSCRIPT
12
CHAPTER 2
LITERATURE SURVEY
2.1 INTRODUCTION
Coated abrasive belt grinding is different from traditional
machining of parts of identical materials. The difference is in the way of chip
production, the order of specific cutting pressure encountered and surface
integrity of machined parts. For better understanding, a detailed literature
survey has been carried out. A resume of the same is presented in this chapter.
The literature survey has been grouped into three subgroups as analysis of
material, wear behavior of coated abrasives and evaluation of abrasive belt on
different work pieces and coolants.
2.2 BELT WEAR, SURFACE FINISH AND MATERIAL
REMOVAL PROCESS IN BELT GRINDING
Abrasive belt grinding is a common finishing process in the metal
and wood working industries. Coated abrasive belts are used in the same
speed range as bonded wheels, but they are not generally dressed when the
abrasive becomes dull (Ernest et al 1990). The wide spread application of
coated abrasive belts and long standing operating practices have led to
different kinds of quality requirements. The abrasive belt machining
technique is more significant for precision machining and finishing, also used
for roughing. They have reported that the performance difference between sol
gel alumina grain and fused grains, it was observed, similar trend on material
13
removal and belt wear for different workmaterials. Though the trend found to
be same but the solgel grain gives more material removal, this is due more
cutting points per grain. Precision grinding gives considerably lower material
removal rates, but can engage large workpiece volume in view of large
workpiece engagement areas possible. Thus wide belt machines can be used
at infeed ranging between 5 and 10 m for economical machining of
workpiece having a width of more than 2000mm (Christian, 1990).
Jourani et al (2005) have reported that the belt grinding improves
the surface finish that was obtained by hard turning. Three dimensional
numerical model developed by them finds that the real contact area and
contact pressure are smaller than the nominal contact area and contact
pressure. This could be due to flexibility in contact. This model was
developed by considering the abrasive grains as distribution of indenters with
various rake negative angles. They have observed that the local normal and
tangential force increase with negative angles. The surface topography and
material ratio are improved with successive processes of hard turning and belt
grinding. They have also reported that the geometrical accuracy reached is
similar to that those obtained using a grinding process. It has been claimed
that it is important to have an idea on the contact pressure, normal, friction
forces and local friction coefficient distribution, to asses the belt wear.
Shibata et al (1979) investigated on wear characteristics of the
abrasive cutting edges on coated abrasive belts by microscopic observations.
The heights of most grain cutting edges on coated abrasive belts gradually
decrease from the tips owing to attritious wear. Therefore, except for a short
initial period when the belt is new, grinding is performed mainly by abrasive
cutting edges with worn grain flats at their tips. The effects of these flats on
belt grinding characteristics were investigated theoretically and
experimentally. A metal removal model was proposed to explain the belt
14
grinding characteristics and their changes with respect to time. The shape and
distribution pattern of the grain cutting edges have more effect on the
performance of an abrasive belt. Hugh Dyer (1955) investigated the
dependency of certain controllable grinding conditions on the wear of
abrasive grains. It was reported that belt speed influences the attritious and
fragmentation wear. They have concluded that high belt speed leads to higher
temperatures at the abrasive grain-metal contact. At low belt speeds
fragmentation wear of the abrasive belt is promoted. It was concluded that the
effective use of abrasive belt was related to the balance between attritious
wear and fragmentation/spalling wear. The wear pattern of coated abrasive
would be different from that of the bonded abrasives. The individual grains
were subjected to unique load. Since the shape of the abrasive grains used for
belt grinding is acicular and also friable, at higher belt speed grain may
getting fragmented.
Mcgibbon et al (1976) have stated that, in high-rate grinding
applications, it was possible to achieve uniform abrasive wear on the coated
abrasive belt. This uniformity of wear permits efficient usage of available
coated abrasive area and aids in producing good flatness tolerance on the
work pieces. In this study, abrasive belts were tested under selected operating
conditions and test results were correlated with predictions.
Abrasive wear of coated abrasives can be compared with the single
grain evaluation. Hamdi et al (2003) have investigated on the cutting power in
a high-speed scratch test device in order to understand the grain behavior and
the wear mechanisms. They have presented two useful and complementary
experimental approaches for understanding the interface physics. An
approach for the specific abrasion energy computation was also presented. It
was reported that the high-speed scratch test for the study of the grain
behavior yield the nearest to the actual results and gives more qualitative
15
information of the grinding process. The experimental results of the grain
behavior presented in the paper were compared with numerical simulation of
the scratch test.
Date et al (2001), have studied on effect of grit size on the initial
performance of fresh coated abrasives and the deterioration of coated abrasive
performance with continued usage. Abrasion tests were performed on pin-on-
cylinder set up which had removable segments for observing the coated
abrasive surface in the scanning electron microscope (SEM). This shows
direct correlation between measurements of coated abrasive performance and
SEM observations of coated abrasive morphology. With coated abrasives
containing finer grit sizes, numerous adhesive wear particles were found on
the coated abrasive surface; this supports the theory that the smaller initial
abrasion rate with finer grits is due to abrasive grains making “elastic” contact
with the metal specimen at loads insufficient for cutting. With continued
usage, the rapid deterioration in performance with finer grits was
accompanied by a buildup of metal caused by capping of the abrasive grain
tips with metal chips and by clogging due to metal chips and adhesive wear
particles becoming stuck between the grains. Coarser grits, were found to
experience extensive grain fracture followed by some grain capping and
flattening but virtually no clogging, the deterioration in coated abrasive
performance was very much less.
Kazuhisa et al (1992), have studied the topography assessment of
coated abrasive tape for distinguishing its functional performance. The three-
dimensional distribution of projecting abrasive grains on a tape surface was
presented and the related geometrical parameters for classifying the coated
abrasive tapes were proposed. A comparison was made between those
parameters of the tape surface and the generated surface textured on the
aluminum alloy substrate. The topographical parameters of textured substrate
16
surface which are based on the profile amplitude. Micro geometry near the
valley bottoms and their number were correlated fairly well with the spatial
distribution of the effective coated abrasive grains in conjunction with
working fluid or coolant.
The individual grain contact with work piece is influenced by the
machine and product parameters. Khellouki et al (2005), have discussed on
the wear mechanism of the abrasive film and its influence on surface
roughness. A theoretical model was developed from the contact conditions
between abrasive film and the surface. Abrasive grains were considered as
cones and wear of the abrasive grain was compared as height reduction of the
cone. The effective contact duration between grains and machined surface and
average contact pressure, were considered as key parameters. The correlation
of key parameters with surface roughness and material removal were
discussed. It was concluded that the effective contact duration and number of
active points are the most important parameters. This model also revealed the
existence of an optimum belt finishing duration depending on the force and
the hardness of the contact wheel. They have reported that material removal
has direct correlation with effective contact duration and hardness of the
contact wheel.
John et al (1985) have studied the coated abrasive belt grinding
through profilometer. The wear characteristics of an alumina and silicon
carbide abrasive were studied. The heights of the abrasives were studied with
respect to time and surface finishes of the workmaterial were compared. It
was concluded that alumina grinds and wears larger than the Silicon carbide.
Shah et al (1977) have described making of computer-controlled
profilometer to describe the quantitative characterization of extremely rough
surfaces such as grinding wheels and coated abrasives. The use of the system
is illustrated by showing the effects of grit size on coated abrasive
17
topography. Duwell et al (1961) has studied the resistance of abrasive grits to
wear and reported that aluminium oxide grits would develop sub surface
crack owing to low hardness and better toughness and these cracks lead to
small fragments of the grit, successive conversion leads to delamination mode
of wear.
Van et al (2010) have studied a new method to evaluate roughness
parameters considering the scale used for their evaluation. Application is
performed for grinding hardened steel with abrasive belts. Seven working
variables are considered through a two-level experimental design. For all
configurations, 30 surface profiles were recorded by tactile profilometry and
rectified by a first degree B-spline fitting before calculating a set of current
roughness parameters. The relevance of each roughness parameter, to
highlight the influencing process parameters, is then estimated for each scale
by variance analysis. The results show that each influent input parameter is
characterized by a related relevant evaluation length.
Khellouki et al (2007), studied the effect of abrasive film and
grinding parameters on the surface roughness. The abrasive film was applied
on work piece with oscillating pad. It was concluded that the belt finishing did
not influence on the waviness parameters and also on the form parameters. Its
action influences only the roughness parameters. Moreover, among the
parameters of belt finishing, it was shown that the granulometry of abrasive
films is the most influential parameter on surface roughness. However, after
the strategic choice of the granulometry, the most influential parameters of
belt finishing are, the applied force, the film feed rate, the hardness of the
contacting roller and the belt finishing duration. The oscillation’s frequency,
oscillation’s amplitude and tangential speed of work piece were shown as
secondary influencing parameters.
18
Mohamed El Mansori et al (2007) have discussed on the belt
finishing process. The influence of the working variables in belt finishing
process like cycle time and axial oscillation frequency on surface finish of
hardened steel was studied. It was concluded that angle of contact, cycle time
and oscillation frequency were significant factors. Axial oscillation of
abrasive band decreases the friction at the grain-microchip interface and
increase in cutting ability by frequent evacuation of microchips and leading to
minimum load.
Meghani et al (2008), have studied the effect of grit sizes on the
surface finish of different workmaterial. A set of parameters was defined
which describe the aluminium oxide resin-bonded belt characteristics
including active grits density, cutting edge dullness, chip storage space and
mean effective indentation. A parametric study was made on the effects of
coated belt characteristics on surface finish performance with different grain
sizes for grinding different workpiece materials. Experimental results are
discussed in relation to the prevailing mechanisms of the process at the belt-
work interface which can be separated into cutting and ploughing components
In coated abrasives, the mineral wear is more critical for a given
bonding system. The abrasive minerals particle size follows a distribution
pattern and the coating process follows the same. The area covered by the
abrasives also should follow a distribution pattern. Burney et al (1975), have
discussed a statistical model for wear of coated abrasive. Statistical model
was created based on the profile height studies. The difference between new
and worn out profiles was evaluated based on auto correlation function,
standard deviation of heights, and standard deviation of profile shape, number
of peaks per unit area in contact and ratio of real to apparent area of contact. It
was also reported that model was developed with the assumptions of coated
abrasive surface as isotropic, contact peaks as circular and heights of the
grains were following Gaussian distribution. It was concluded that abrasive
19
wear in the belt grinding resulted in slower decay in auto correlation,
continuous reduction in the standard deviation of profile heights, profile slope
and profile curvatures. It was reported that belt wear showed reduction in the
number of peaks per unit contact area. This means that abrasive wear is
relatively controlled.
Phadke et al (1975), have studied the coated abrasive profiles using
second order continuous autoregressive model. Expressions were derived
from the geometry of the average grain size. These models were developed to
understand the abrasive wear. It was concluded that the average grains
become shaper as its size reduces. As per the model, it was observed that apex
angle increases with the increase in grain wear out.
The distribution of abrasive grits in grinding tool plays a critical role in
deciding the surface finish. Andreas et al (2003) studied the feasibility of
using engineered abrasives for consistent surface finish. Recent advances in
engineered abrasives have allowed replacement of the random arrangement of
minerals on conventional belts with precisely shaped structures uniformly cast
directly onto a backing material. This allows for abrasive belts that are more
deterministic in shape, size, distribution, orientation, and composition. A
computer model based on known tooling geometry was developed to
approximate the asymptotic surface profile that was achievable under specific
loading conditions. Outputs included the theoretical surface parameters, Rq,
Ra, Rv, Rp, Rt, and Rsk. Experimental validation was performed with a custom-
made abrader apparatus and using engineered abrasives on highly polished
aluminum samples. Interferometric microscopy was used in assessing the
surface roughness. It was concluded that the individual parameters like
pyramid base width, pyramid height, attack angle, and indentation depth on
the surface have quite a strong relation with surface roughness.
Mulhearn et al (1962) have discussed on the abrasive finishing
technology which employs a soft coated belt as a tool; an attempt was made to
20
identify the simultaneous influence on the process efficiency of two working
parameters: the contact pressure and the abrasive particle size. Their effects
were mainly studied by considering the most important achievements of the
belt finishing process which is to reduce the workpiece surface irregularities
and to improve its geometrical quality with a lower wear of the abrasive belt.
Chang et al (2003) have discussed on the finishing tests done in wet
conditions with various contact pressure and average size of Al2O3 abrasive
particles. With all other working parameters kept constant, five values of
contact pressure (between 0.3 and 0.8MPa) and six abrasive belts of different
grains size (9, 15, 30, 40, 60 and 80 µm) have been considered. It was
concluded that contact pressure range has effective correlation with the rough
super finishing range.
Grzesik et al (2007) studied the process of hard turning followed by
the abrasive machining. This is because, many transmissions parts, such as
synchronizing gears, crankshafts and camshafts require superior surface finish
along with appropriate fatigue performance. Belt finishing process was
performed with pressure and oscillation. In this investigation, 2D and 3D
surface roughness parameters, as well as profile and surface characteristics,
such as the amplitude distribution functions, bearing area curves, surface
topographies and contour maps were monitored and analyzed. Experimental
data collected during measurements indicate that each of the finishing
abrasive processes provides a specific set of surface topologies. The
transformation of bearing properties of surfaces, generated through two
optional hard turning-BG and MC HT-SF machining sequences, were
highlighted. As a result, the modifications of surface profiles achieved by
means of special abrasive machining operations, distinctly improved the
bearing properties of previously hard turned surfaces, and exemplarily, they
shorten the running-in period. It was concluded that elastic belt modifies the
valleys and peaks of the surface. The modification of hard turned surfaces by
abrasive finish, changes the partition of height and spacing roughness
parameters in the total profile height Rz. They have found that hard turned
21
surfaces treated with belt finishing give better bearing properties than by rigid
abrasive stone.
Hong et al (1975) have studied the effect of contact wheel on
centerless belt grinding. Different kinds of contact wheel and its impact on
output parameters were studied. It was concluded that metal base contact
wheel gave better belt efficiency and dimensional accuracy on the finished
part. Mineral utilization was found to be higher for metal contact wheel
whereas the rubber contact wheel resulted in 35 -65% utilization owing to
enhanced elasticity and elastic contact area.
Recha et al (2008), studied on the methodology on AISI 52100
work material. Belt grinding process was performed to reduce the residual
stresses generated in the hard turning. It was shown that the belt finishing
process improved very significantly the surface integrity by the induction of
strong compressive residual stresses in the external layer with significant
improvement of the surface roughness. A two stage process was discussed for
belt finishing and in the first step belt eliminates very quickly the peaks of the
surface texture until abrasive grains reach the lower part of the surface
texture. Further to reach of lower part, the shape of the surface texture
remains constant. Second step was defined as rubbing phase where the
abrasive grains are rubbing the work material. Work material was ploughed
on each side of the grains. It was of the effect on subsequent layer was
influenced by the pressure on unit grain. concluded that the compressive layer
affected by the belt finishing was influenced by the lubrication and lack of
lubrication would induce tensile stress.
Fine grit abrasive belts are used in the finishing process. The
material removal in the finishing process is very low. The scratch pattern on
the surface decides the finish, hence the shape and size of the abrasive plays
critical role. Any grinding parameter which decides the contact points
between work piece and abrasive should influence the finishing process.
Mezghani et al (2009), have studied the effect of grit size on the finishing of
22
cast iron and steel components. Geometrical and superficial variations of the
workpiece samples were measured in 2D surf scan and surface profile of the
abrasive grain morphology were measured after removing the superficial
microchip layers. It was reported that predominant mechanism for finishing
depends on the size of abrasive grains and the wear state of abrasive belt.
Height reduction in the grit changes the shape and morphology of the grit
which changes the attack angles of finer abrasive grits. It was reported that
these changes were favoring the plowing mechanism. Junji Sugishita et al
(1978), have discussed the wear process of silicon carbide paper. Results were
obtained in the investigation of 600 and 100 grade water proof papers.
Abrasive paper was studied in different pressure and reciprocating load. The
rate of detachment of grits, the material removal rate and the mean sizes of the
detached grits were investigated under various conditions. Experimental
results were reported that reciprocating frictional action created more abrasive
wear, possibility due to stress attributed over the peaks of abrasives. It was
concluded that the mean sizes of the detached grit were independent of load
differences and speed differences in the strokes and the wear of cutting points
was high at the initial stage of the testing.
Ren et al (2007), have discussed the process model to estimate the
material removal rate in the robotic belt grinding. Finite element analysis was
used to simulate the belt grinding process. Abrasive grit size and shape were
included as one of the parameters. Model was developed by considering three
dimensions namely contact situation, force distribution and removal
computation. The contact situation was related to geometric intersection
between grinding belt and workpiece. This was used to calculate the force
distribution. The proposed model included the geometry of the work piece as
the one of the parameters.
Xiang Zhang et al (2004) have discussed new model based on
support vector regression (SVR). This model was relating the nonlinear
relation between the local contact situation and force distribution. Xiang
Zhang et al (2005) have developed a local grinding model to simulate the
23
robot-controlled belt grinding processes, especially for grinding free-form
surfaces. A new force model to estimate the grinding forces is also put
forward as an alternative to the conventional FEM model. Instead of handling
the problem in pure physical way, the new model indicates the nonlinear
relationship between local contact situation and force distribution. Bigerelle
et al (2009), have discussed on surface finish by belt grinding process. Scaling
analysis was used for roughness characterization. It was concluded that
roughness amplitude has direct correlation with scan size and belt finishing
process creates a fracture structure on tooled surfaces until a critical length
which is related to the profile autocorrelation length. It is informed that larger
contact area and enhanced uniform compatibility in belt grinding results in the
above findings. Yuko et al (2005) studied on the surface finish. The surface
finish of the samples as shown in Figure 2.1 were sanded using various grades
of coated abrasives and the roughness parameters, such as reduced peak
height (Rpk), core roughness depth (Rk), and valley depth (Rvk), were
estimated on the material ratio curves, which were obtained from roughness
profiles determined using robust Gaussian regression filter.
Average Peak-to-Valley vs Material removal
0
2
4
6
8
10
12
14
0 5 10 15 20 25 30
Material removal x 1000 cm3
Av
era
ge
Pe
ak
-to
-Va
lle
y in
m
Figure 2.1 Average peaks to valley vs material removal
(By volume)
24
Li Na Si et al (2009), have carried out a study on optimization of
belt grinding of fiberglass reinforced plastic. Experiments were performed on
wet abrasive belt grinding of glass reinforced plastic, the reasons for belt
slipping was analyzed. It was concluded that the hardness of the contact
wheel plays a significant roll. Abrasive belt grinding technology has been
applied in most material machining processes with its high machining
efficiency low grinding temperature and high machining precision. Zhi Huan
et al (2009), have studied on abrasive grain granularity, belt speed and
workpiece feed. Speed plays an important role in grinding for magnesium
alloy tube surface, magnesium alloy tube surface roughness (Ra) ranging
between 0.22 um to 2.93 um. The belt grinding was studied as pre treatment
process for magnesium alloy which decides on the surface texture /pattern of
magnesium alloyed components. It was concluded that increasing belt speed
and workpiece feed deteriorates roughness.
Kayaba et al (1986) have discussed on the effect of contact pressure
in belt finishing operation. It was reported that applied contact pressure
modified the abrasive efficiency of coated belt. It was concluded that wear
rate of abrasive grits were depends Workpieces/tool contact conditions and
significance of the contact pressure varies with abrasive grains size. Visser et
al (1981) have reported that in belt finishing process, under wet condition, the
adhesion between aluminium oxide abrasive and workpiece microchips is
negligible. Zhi Ming Lv et al (2009), have discussed on the coated abrasive
belt finishing of slender piston rod. The belt grinding process was compared
against conventional grinding wheel process. It was concluded that surface
finish in belt grinding was improved surface compatibility due to elastic
contact point of belt grinding.
Hyunsoo Kim et al (1990) had devised an equation based on Euler
formula and investigated the normal and tangential belt force distribution in a
25
flat belt grinding. The belt force equation presented in this paper has a
constant coefficient of friction. Analytical results showed good agreement
with the experimental results and the authors' previous work, which was based
on a surface model of the belt friction force with a varying coefficient of
friction. The equation developed from the Euler formula was suggested for
practical design due to its sufficient accuracy and simplicity.
Robert et al (1980) have discussed on the wear and performance
characteristics of two abrasive minerals used in two different modes of belt
grinding. The modes are low pressure contact-load and high pressure constant
rate. The abrasives such as alumina and alumina-zirconia were compared for
their wear characteristics .The belt caliper, surface profile were studied to
understand the grinding performance with respect to pressure. It was reported
that profile of the grains on a new coated abrasive was distributed over a wide
range of heights. They have reported that at mild constant load test only a few
high grains were participating in the material removal and at severe constant-
rate test more number of grains was contacting the work pieces. Belt caliper
curves for different mineral type for same grinding mode are more alike than
curves for belts of the same mineral type used in different grinding modes.
The amount of minerals consumption changed with respect to loading pattern.
It was concluded that alumina zirconia abrasive removes ten times more
material than by alumina at constant load test. This is possibly due to
enhanced friability or toughness of the abrasive grain.
In coated abrasive, it is possible to change the wear pattern based
on the product design. A protective layer shall create the differences in the
wear mechanism. Billingham et al(1974) studied the grit failure mechanism in
coated abrasives belt grinding on a wide variety of engineering materials.
Study was related to the effect of additional “size” layer on the product with
“anti-glaze” property. Scanning electron microscopy was used to both
26
characterize and monitor the incidence of the various grit failure mechanisms
for a standard aluminium oxide resin-bonded belt and a similar belt
containing a grinding aid additive. The two predominant wear mechanisms
were discussed, namely capping, where swarf becomes firmly attached to the
grit surface preventing any further grinding action and by dulling, which is a
combination of attrition by chemical degradation or plastic flow and small-
scale grit fragmentation, which leads to the formation of flats on the grit
surfaces. It was reported that antiglaze layer has reduced the belt wear. Wear
due to capping was reduced due to the antiglaze layer. The effect of anti-
glaze found to be the same for different work pieces like stainless steel, super
alloys and high carbon steel.
The wear of coated abrasive shall be correlated with single grit
grinding. Doyle et al (1978) reported that delamination of the surface layers
of brass work pieces was observed during single-grit grinding. In most
instances, complete delamination did not occur and the delaminated material
was still attached to the workpiece. Transmission electron microscope
examination of the ground surface layers showed that they had a fine sub
grain structure; no evidence of a dislocation free surface layer was obtained,
Delamination was found to occur in the fine grained surface layers, and it is
suggested that this occurred by a process of shear separation within the fine
sub grain structure. The manner in which the delaminated sheets were
observed to lift away from the workpiece surface seems to indicate the
presence of both residual tensile and compressive stresses within the surface
layers.
Desa et al (2001) stated that in machining of ceramic materials
subsurface damage and low material removal rates was observed, owing to
their high hardness, low thermal conductivity and brittle behavior. With this
in mind, the subsurface damage and material removal in single point
27
scratching and abrasive machining of alumina and silicon nitride were
studied. The lubricants were selected based on the potential for high material
removal rates and low subsurface damage as determined from an earlier
study. A Vickers pyramidal indenter with a linear transverse motion was used
for scratch tests while abrasive machining was done under controlled load
conditions in a modified belt sander. The material removal rates in scratch
tests were determined by profilometry and in abrasive machining by
gravimetric measurements. The subsurface damage was studied by scanning
electron microscopy (SEM). It was found that the subsurface damage in both
processes was greater in alumina than in silicon nitride. Compared to alumina,
silicon nitride is tougher and able to resist delamination and related damage.
The lubricants were found to contribute to higher material removal rates and
lower subsurface damage as compared to the dry condition. X-ray
photoelectron spectroscopy (XPS) analysis of the abrasively machined
surfaces of alumina revealed that the compounds such as Al(OH)3 and
MeSiO5 were generated on the cutting surfaces. These compounds contributed
to the above effects by making the cutting conditions less aggressive
Masatoshi et al (1983), have discussed on the surface
characterization of coated abrasive grinding of carbon steel. It was concluded
that the distribution of the cutting edges could be represented as limited
parabolic type. Average mean value of the grain size has exponent relation
with surface finish. Bhattacharyya et al (1975) have discussed on the
topography of the coated abrasives. The surface profiles were tested with
stylus measurement technique with subsequently analyzed with statistical
model for output signal. Leggett et al(1978) discussed on the process
parameters for belt finishing and recommended that power source, drive or
contact wheel, an idler for tensioning, tracking the belt, and properly selected
coated abrasive belt were important factors to get good surface quality. These
components differ for all the methods but in general the workpiece is pressed
28
between the grinding head and the rest support. The objective of the
regulating head is to coordinate the belt pressure. Wang et al (2008), have
discussed on the selection of belt grinding axis on finishing of turbine blades.
Blade grinding was realized by two liner control axes and two rotation control
axes. The kinematic model was developed and reported that belt grinding as a
suitable process for turbine blade finishing. Konig et al (1986), have
discussed on the belt grinding parameters like contact wheel and surface
qualities of the workmaterial. They have concluded that for high surface
qualities the contact wheel hardness, axial accuracy and area of contact are
critical parameters. It was reported that axial accuracy decreases with
increase in the hardness of the contact wheel. Harder contact wheel reduces
flexibility over the contact region.
2.3 BELT GRINDING WITH COOLANTS
One of the important advances in the coated abrasive field since
development of the synthetic abrasive has been the development and
utilization of special grinding aids. These in the form of fluids and waxes
which permit the use of coated abrasives for economical grinding and
polishing. The difficult to grind materials such as plastics and super alloys
finished by belt grinding and by the addition of grinding aids. Since the
effective contact area of the coated abrasives and heat generation in the
grinding process are small, it needs more optimization to create effect of
external atmosphere in the belt grinding process. Amundson et al (1975) have
studied the influence of aqueous solution of K3P04 on material removal rate in
coated abrasive machining. It was reported that 10% aqueous solution of
K3P04 buffered with NaH2PO4, act as grinding aid. In grinding of 1095 and
AISI 304 stainless steel the material removal rate decreases with increase in
the cut per path. The maximum power drawn in grinding was found to be low
in dry grinding compared to wet grinding. It also concluded that abrasive
29
wear was decreased due to wet grinding which inturn reflected as increase in
material removal. Reduction in wear was related with adhesion between
mineral grains and metal. It was also reported that mineral metal interaction
creates micro fracture in the mineral and mineral pullout from the backing. It
was referred that K3PO4 solution worn down the grain attritiously to a less
sharp.
Mercer et al (1988), have discussed on the influence of atmospheric
composition such as air, inert gas, on the abrasive wear of coated abrasive
paper on grinding of titanium and Ti-6A -4V. Wear and frictional behavior of
metals on silicon carbide and alumina were studied in pin-on-disc method.
The abrasive wear observed was grit blunting and build-up-edge or capping.
Yu Fu Wang et al (2009), have discussed about grinding of titanium alloys.
These alloys are sensitive to heat generated at grinding zone. Grinding
process creates, heat affected zone and burn marks which result in
degradation of mechanical and metallurgical properties. It was stated that belt
grinding has more sharp cutting points and pointed grains give sufficient chip
clearance. This paper attempts to retain the pointed structure of belt and
improve the material removal rate by coolants. They have recommended the
green cooling technology by liquid Minimum quality level (MQL). It was
reported that liquid nitrogen MQL significantly influences belt wear and
material removal rate. Liquid nitrogen acts like a grinding aid which reduces
the friction coefficient and toughness of the work materials. Liquid nitrogen
easily access to grain-workpiece interface and reduces the coefficient of
friction, and hence less damage, reduced build-up.
Heat sensitive materials can be finished with coated abrasive belt
grinding. The wear pattern changes with grinding time. Cadwell et al (1961)
have studied the factors affecting the performance of fully and conventionally
coated abrasive belts in grinding of the Ti- 6 A1- 4V. The decline of
30
volumetric stock removal by attritional wear of the abrasive wear was studied
by an automatic sensing and plotting methodology. It was reported that the
effect of lubrication varies with respect to surface speed of the belt.
Experimental studies on lubricants such NaNO2 and K3PO4 , showed that
greater mobility of nitride ions made the effect of belt speed less significant
on wear. Axinte et al (2005), have worked on optimization of belt polishing
for Ti-6Al-4V alloy. Belt polishing was employed subsequent to milling
process. The belt speed, pressure and table speed were used as variable
parameters. The structured abrasive belt was used in a finishing operation.
The structured abrasive was considered as belt with uniform height of the
cutting points. It was reported that belt finishing of Ti-6Al-4V generated very
less heat. The surface integrity of the workpiece found to be same for dry or
air lubricated, air cooled grinding. The life of the belt was evaluated against
the optimal measures such as surface roughness, material removal and
machining time. Axinte et al (2009), have discussed on the polishing of Ti-
6Al-4V heat-resistant alloy. Two types of polishing methods namely belt and
bob polishing methods with various media/grades (Al2O3, SiC, polycrystalline
diamond) of the abrasive materials in conjunction with three cutting media
have been tested to address the overall finishing of components. It was
discussed that both polishing methods are capable to meet the requirements of
minimum workpiece surface coverage. When considering surface roughness
criteria, Al2O3 belts and SiC bob tools were found appropriate. Furthermore,
surfaces obtained with these tools when employing cooling media (chilled air
for belt polishing and minimum quantity of lubricant (MQL) for bob
polishing) showed compliance with the tight requirements of industrial
standards for workpiece surface integrity (metallurgical damage and residual
stresses). This proved that belt and bob polishing methods can be employed to
enable automated overall finishing of complex geometrical components
31
Mamoru et al (1987), have reported the optimization study on belt
grinding of stainless steel. In this study on formulation or selection of an
optimum oil-based grinding fluid with which stainless steels can be
successfully ground, an optimum base oil was first experimentally selected,
and then additives were evaluated for their effect in improving abrasive-belt
grinding performance. Chemical grinding oil additives were found markedly
effective in improving the abrasive-belt grinding performance for both 19Cr-
2Mo ferritic stainless steel and SUS 304 austenitic steel, with those containing
sulphur or chlorine being superior to those containing phosphorus, fatty acid
or alcohol. Among all the additives tested, chemically active oils, such as
sulphurized mineral oil, exhibited the best performance.
Nakayama et al (1988), have studied the effect of blended paraffin
base oils of the same viscosity on the abrasive belt grinding of stainless steel.
The base oils are varying in specific gravity, viscosity and sulphur content.
Oil was applied through pressurized cylinder. The effect of viscosity and
blending ratios on material removal and belt wear were reported. The
following conclusions were reported in the study. Stock removal was
decreasing with increase in oil viscosity. Abrasive belt wear was decreasing
with increasing viscosity. Fracture wear was observed at low viscosity and
abrasive wear at higher viscosity. At low viscosity oil film was easy to
penetrate hence, more wear was reported. The optimum lubricant was
measured based on grinding ratio i.e. the maximum material removed at
minimum belt wear. Maximum Grinding ratio was reported with blended oil
lubricant.
Hugh (1955) studied the the life of abrasive belt in different
lubrication environment in griniding of carbon steel 1020 and stainless steel
304. Figure 2.2 shows the results of tests using aluminum oxide of grit size
50, resin bonded, cloth backed, belts to grind both low carbon steel (1020 hot
32
rolled)and stainless steel (Type 304) under the same conditions of high unit
work pressure. The belt speed was 25m/s, and the metal test piece was
oscillated across the belt at a speed of 0.35m/s under a constant pressure of
5kg.
Figure 2.2 comparative stock removal-accelerated shedding tests
It was reported that reduction of belt life was often the result of
changes in one or more of the above listed variables. They have concluded
that it was required to adjust the appropriate variables to reach the ideal
situation for each job.
Wilke et al (1986), have studied the coated abrasive super finishing
of metal roll surfaces. Coated abrasive tape was used as finishing tool, belt
speed; oscillation and pressure were used as parameters. They have studied
the effect of grinding atmosphere dry, water and oil on the performance of
coated abrasive belt finishing. The water soluble synthetic lubricant used to
flush debris from the interface was found to enhance abrasive action. Plain
water, mineral seal oil and other lubricants were tested. Use of mineral oil
1020 Dry
1020 Oil
1020 Emulsion
304 Dry
304 Oil
33
resulted in coarse surface finish on hard workmaterial. It was reported that
coated abrasive super finishing conducted in the absence of lubricant caused
the abrasive tape surface to become loaded.
2.4 GRINDING OF BRITTLE MATERIAL
Hard and brittle materials like alumina, silicon carbide, refractory
materials are finished only with super abrasive grinding wheels. Since
grinding forces are high and abrasive wear rate is also high. There is only few
research works have been carried out on coated abrasive belt grinding of
ceramic materials. Wang et al (2003) have studied the platen belt grinding of
brittle materials like ceramic tile, granite and marble. They have conducted
the tests with silicon carbide and zirconia base abrasive belts on vertical
loading conditions. The effect of belt speed, type of abrasive, force and
coolant on abrasive belt grinding at constant pressure were discussed. It was
reported that abrasives were able to grind granite and marble in reciprocating
and rotating worktable. It was concluded that belt grinding had no significant
effect on ceramic tiles. This could be due the glazed surface of the ceramic
material. The grinding conditions were optimized, with grinding pressure,
worktable rotation, belt speed and coolants. It was reported that worktable
rotation had greater influence on material removal and surface finish. It was
concluded that belt speed reduce the cutting force at higher belt speed.
Yoangsheng et al (2003), studied the abrasive belt grinding of
granite. Based on analysis of action of surfactant in granite grinding with
diamond disk, the effects of the grinding coolant with different surfactant and
concentration were studied in belt grinding. The influences of different
parameters like grit size, cationic, anionic and non ionic surfactants on belt
grinding were reported. It was reported that material removal rate improved
greatly using grinding coolants with surfactant. Addition of Surfactant is
leading to reduction in the surface formation energy and hence increases in
34
material removal. The grinding speed, grit mesh size and rotating table speed
were affecting the function of surfactant. The belt speed 15 mps and the work
table rotation of 50 rpm was reported parameters. Addition of surfactant leads
to reduce the surface formation energy and consequently increases material
removal. Anne venu gopal et al (2004) investigated that the chip-thickness
model, used to assess the performance of grinding process, plays a major role
in predicting the surface quality. They have concluded that the effectiveness
of the existing chip-thickness model has been enhanced by incorporating
elastic properties of the wheel and the workpiece. This model exhibits the
importance of incorporating the material properties of the wheel and
workpiece apart from other traditional grinding parameters. The new chip-
thickness model was more accurate in predicting the surface roughness
Sanjay Agarwal et al (2007) reported that it was often desired to
increase the machining rate while maintaining the desired surface integrity.
The success of this approach, however, relies on the understanding of
mechanism of material removal on the micro structural scale and the
relationship between the grinding characteristics and formation of
surface/subsurface machining induced damage. In this paper, grinding
characteristics, surface integrity and material removal mechanisms of SiC
ground with diamond wheel on surface grinding machine have been
investigated. The surface and subsurface damage have been studied with
scanning electron microscope (SEM). The effect of grinding conditions on
surface/subsurface damage has been discussed. This research links the surface
roughness, surface and subsurface damages to grinding parameters and
provides valuable insights into the material removal mechanism and the
dependence of grinding induced damage on grinding conditions.
Gowri et al (2006) have investigated on “Modeling and
Optimization of super abrasive grinding of alumina ceramics”. It is stated that
35
super abrasive grinding of advanced ceramics requires effective selection of
process parameters to control the surface integrity and to maximize the
material removal rate. Grinding of advanced ceramics such as alumina is
difficult due to its low fracture toughness and sensitivity to cracking.
Xipeng Xu et al (2003), have studied the mechanisms of abrasive
wear in the grinding of titanium (TC4) and nickel (K417) alloys. The present
investigation was dedicated to elucidate abrasive-wear mechanisms during
surface grinding of a titanium alloy (TC4) and a nickel-based super alloy
(K417) using silicon carbide (SiC), alumina (Al2O3), and cubic boron nitride
(CBN) wheels. The temperature at the wheel–workpiece contact zone was
measured using a workpiece–foil thermocouple. SEM and EDS were used to
examine the morphological features of ground workpiece surfaces and worn
wheel surfaces. It was shown that the grinding with either SiC or Al2O3 is
characterized by the high temperatures reached in the grinding zone since
either of them is easily worn during the grinding processes. Along with the
presence of high temperatures, strong adhesion was found between the
abrasives and workpieces, which might be attributed to the chemical bonding
between the abrasives and workpieces at the elevated temperatures.
Increasing ductile deformation of both TC4 and K417 at the elevated
temperatures may also be a factor. Therefore, the wear of SiC or Al2O3 is both
chemical and physical. In the grinding with CBN wheels, however, the wear
of abrasive grits is mainly physical since CBN is more stable at higher
temperatures. At extremely high temperatures, CBN was found to undergo
dislodging prior to being gradually worn. In order to reduce the grinding
temperature, a segmented wheel was incorporated into grinding with CBN
wheels.
Kun Li et al(1996) reviewed the published works related to
surface/subsurface damage and the fracture strength of ground structural
36
ceramics over the last 20 years. Consistent as well as conflicting experimental
observations concerning grinding-induced micro cracks, residual stresses and
the degradation of flexure strength are summarized. Special attention was
shown on the quantified relationships between the grinding variables,
machining damage and the flexure strength of ground ceramics because of the
importance of these relationships for optimizing the ceramics grinding
processes. Several issues such as non-uniformity of the grinding interface,
crack measurement, thermal effects and micro structural features are
discussed and highlighted as future research areas for better modeling and
simulation of ceramics grinding operations.
Xipeng Xu et al (2003) in their study measured the temperature at
the wheel–workpiece contact zone using a workpiece–foil thermocouple.
SEM and EDS were used to examine the morphological features of ground
workpiece surfaces and worn wheel surfaces. In order to reduce the grinding
temperatures, a segmented wheel was incorporated into the grinding with
CBN wheels. Duwell et al (1988), have studied the performance of coated
abrasive lap covers with free abrasive lapping. The work materials studied
were fused quartz, pyrex, alumina and silicon nitride. Above studies were
conducted in dry conditions and with oil and water. It was concluded that
water was highly beneficial to the performance of coated abrasive on all the
materials. They have reported that conventional minerals in coated form may
improve the surface integrity in the final finishing of glasses, ceramics by
avoiding subsurface crack formation. They recommended that high normal
load or coarse grades were to be avoided for finishing of ceramic
workmaterial.
2.5 SCOPE AND OBJECTIVE OF THE PRESENT WORK
The literature survey on abrasive belt grinding mainly indicates the
applications in the non-precision finishing processes. Figure 2.3 shows the
37
differences between belt grinding and finishing. There was considerable work
done on belt finishing process. Most of the studies were restricted to the low
pressure and low belt speed grindings. Studies conducted on final finishing
were assisted with work pad oscillation vibrations and belt grinding was
studied as one of the finishing process. In the belt finishing process, studies
were oriented towards the reduction of residual stresses, created in the
previous processes. Yet definite conclusions could not be arrived. Not much
on belt properties and relative influence on machining parameters is reported.
Regarding the influence of grinding environment no conclusive information
can be arrived and role of lubricant and grinding fluids is yet to be
understood. There are continuous developments in the abrasive mineral,
bonding system, backing materials, manufacturing process and workpiece
materials, which demand study on influence of belt characteristics and
machining parameter on the output of belt grinding. Hence there is a need for
further study in abrasive belt grinding for proper understanding of the
parametric influence and also to develop useful data base.
Figure 2.3 Belt grinding and finishing methods.
38
2.5.1 Need and Scope
The literature survey on belt grinding shows certain limited
understanding of material removal, wear and grinding process. The
importance of belt related parameters in grinding and finishing of workpieces
can be seen in the illustration on grinding Figure 2.3. Compared to the
grinding with wheels, involving non rigid wheel with belt grinding is another
way to enhance the flexibility. The aim is through systematic approach to
optimize parametric setting to achieve the desired output and precision in
coated abrasive belt grinding.
Accordingly the objectives of the study are to:
Conduct a detailed study on grinding characteristics of coated
abrasive belt grinding process and develop a methodology to
maximize the output and usage of belt grinding.
Conduct experiments to study the belt properties and grinding
parameters to arrive a systematic process to select belt and
grinding parameters.
Analyze the data and develop statistical model considering
individual and interactive parametric influence on performance
indicator.
Recommend the feasibility of optimizing parameters for custom
specific applications.
39
2.5.2 Organization of Research Work
The organization of research work based on the objective is illustrated in
Figure 2.4.
Systemized innovative approach in optimization of coated
abrasive belt grinding
Machining Parameters
1. Belt speed
2. Land to groove ratio
of contact
3. Grinding pressure
4. Grinding
atmosphere
Modeling and Optimization of belt grinding
Grinding Responses: Material removal, Belt wear, Surface roughness and Surface damage
Work Material
AISI 304
Stainless steel
EN08 steel
HCHCr and
Alumina
Abrasive belt
Bonding system
- Glue over Glue
- Resin over Glue
- Resin over resin
Grit size
- grit 24, 36 and 60
Flexing pattern
- 90 degree
- 90+45 degree
- 90+45 -45 degree
Backing material
- cotton
- polycot
- polyester
Results and Discussion
Conclusion
Study on parametric influence on grinding performance
Figure 2.4 Organization of research work